There is an ongoing evolution in scholarly communication. The traditional way of communicating scholarly results through scientific journals is beginning to be replaced by other means of dissemination. One new channel for scholars to communicate their results is digital publication in so-called open archives. An open archive is a kind of full text database, where a scholar may archive/publish his or her articles, as a complement or a substitute to publishing in scientific journals. There are discipline-oriented archives and institutional archives. The Open Archives Initiative OAI, set up in 1999, is a major development aimed at heightening interoperability between different open archives. The aim of this thesis is twofold. The first aim is to give an overview of the development of the OAI, mainly from a knowledge organizational perspective. The second purpose is to examine open archives from a historical, social and cultural perspective. The open archives movement has not been the immediate success that many of the advocators had hoped. It is also said that scholars in the humanities and the social sciences are more reluctant to self-archiving than scholars in the natural sciences. This thesis try to find out more about why scholars are reluctant and to examine whether it is true that scholars in the humanities and the social sciences are more reluctant, and if its true, why this is so. This study compiles some weaknesses with the OAI retrieval model that have been observed lately. Many of these weaknesses are related to human aspects of the retrieval model, for example issues related to the creation of metadata. The results show that scholars in the humanities and the social sciences have been slow to accept open archives, as a reliable channel of publication and dissemination and that the reasons for this may be historical, social and cultural. The study compiles different barriers that may prevent or delay a more widespread usage of open archives. Some barriers may affect scholars in all disciplines; other may mostly affect scholars in the humanities and the social sciences. This study proposes that historical, social and cultural issues have to be taken into consideration when new information retrieval systems are developed.